Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/foundryside-dev/plainweave/claude-mdgit clone --depth 1 https://github.com/foundryside-dev/plainweaveWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.02773 | $0.02773 |
| Opus 5 | $0.01386 | $0.01386 |
| Sonnet 5 | $0.00555 | $0.00555 |
| Haiku 4.5 | $0.00277 | $0.00277 |
Grade C, and why
plainweave CLAUDE.md scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- filigree:instructions:v3.0.1:65e6fb25 --> This is a copy
100% identical to plainweave AGENTS.md — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 201 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Filigree Issue Tracker
filigree tracks tasks for this project. Data lives in .filigree/. Prefer
the MCP tools (mcp__filigree__*) when available; fall back to the filigree
CLI otherwise.
Workflow
# At session start
filigree session-context # ready / in-progress / critical path
# Pick up the next startable issue (atomic claim + transition into its working status)
filigree start-next-work --assignee <name>
# ...or claim a specific issue
filigree start-work <id> --assignee <name>
# Do the work, commit, then
filigree close <id>
Use the atomic claim+transition verbs — work_start / work_start_next
(MCP) or start-work / start-next-work (CLI). Do not chain
work_claim (MCP) or filigree claim (CLI) with a subsequent status
update — the two-step form races against other agents; the combined verb is
atomic.
Ready ≠ startable. The working status is type-specific (tasks →
in_progress, features → building). Bugs start at triage, which has no
single-hop transition into work (triage → confirmed → fixing), so a triage
bug is ready but not directly startable: work_start on one returns
INVALID_TRANSITION naming the next status, and work_start_next skips it.
work_ready items carry a startable flag (plus a next_action hint when
false). Pass advance=true (MCP) / --advance (CLI) to walk the soft
transitions to the nearest working status automatically.
Observations: when (and when not) to use them
observation_create is a fire-and-forget scratchpad for incidental defects — things
you notice outside the scope of your current task (a code smell in a
neighbouring file, a stale TODO, a missing test for an edge case you happened
to spot). Notes expire after 14 days unless promoted. Include file_path and
line when relevant. At session end, skim observation_list and either
observation_dismiss or observation_promote for what has accumulated.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 201 lines · 2,773 tokens per session scan C d0af15e1e5b2
plainweave CLAUDE.md is an instructions file published in the GitHub repository foundryside-dev/plainweave (0 stars, last pushed 1mo ago), licensed MIT. It adds 2,773 tokens to every session, about $0.0139 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). It is 100% identical to plainweave AGENTS.md, differing in 0 lines, and is treated as a copy.
Other instructions, from other repositories
librosa AGENTS.md
Instructions for librosa/librosa, covering agent interaction policy for librosa, 1. automated platform interactions, 2. exception for project-authorized automation and 3. machine-readable directive.
MemMachine AGENTS.md
Instructions for MemMachine/MemMachine, covering memmachine agent guide, repository orientation, environment setup, build, lint, and test commands and python (repo root).
dify AGENTS.md
AGENTS.md instructions for langgenius/dify, covering agents.md, repository gotchas and frontend workflow.
dify CLAUDE.md
Claude Code instructions for langgenius/dify, a project described as: Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
vibe-kanban AGENTS.md
Instructions for BloopAI/vibe-kanban, covering repository guidelines, project structure & module organization, crate-specific guides, managing shared types between rust and typescript and build, test, and development commands.
minion CLAUDE.md
Instructions for femto/minion, covering 🧠 记忆存储, 系统架构记忆 and 开发流程记忆.